Skip to content
deconstructed
Under the Hood · 01

What Actually Happens When You Type a URL Into Your Browser?

A web address looks like one thing, but a browser has to turn it into a conversation across several systems. Follow a typical navigation from the address bar to the pixels on your screen.

Playground · the first two steps

Choose your level

Your browser knows a name. It needs a place.

Computers find each other using numbers, but people remember names. Before a website can open, your browser has to look up the number that goes with the name you typed. Watch that happen, then try making it go wrong.

Ready

You typed the name shop.example, but it needs the matching number before it can connect. Press Next step and follow the question.

How long it took

259 ms

  • Between you and the helper
  • Top of the directory
  • Ending directory
  • The site's own list

Every coloured block is one trip to ask someone. Fewer trips means the page starts loading sooner. A millisecond (ms) is a thousandth of a second.

You can edit this and try any website address.

The parts of an address

scheme
httpsHow to talk to the site. https means a private, secure conversation.
host
shop.exampleThe site's name. Only this goes to the lookup.
port
8443A door number on the site's computer. Usually hidden.
path
/productsWhich page you want, like choosing a room in a building.
query
color=greenExtra details for the page, such as a filter or a search.
fragment
detailsA spot inside the page. It never goes to any server.

Quick experiments

Has this name been looked up before?
Does something go wrong?

Words you just met

DNS
The internet's directory. It turns names people can remember into numbers computers use.
IP address
The number that says where a computer is on the internet, like 192.0.2.5.
Resolver
The helper that does the asking on your behalf and remembers answers for a while.
Cache
A saved copy kept close by so the same question doesn't need asking twice.

This is a simplified model made for learning. The timings and the 192.0.2.x numbers are examples, but the steps are how the internet really finds a website. The rest of the journey continues in the written walkthrough below.

The whole journey in seven lines

A typical new HTTPS page load. Caches, redirects and parallel work can skip or reorder some of this.

  1. 01

    The browser parses the URL

    The browser separates the scheme, host, port, path, query, and fragment.

    https://shop.example:8443/products?color=green#details

  2. 02

    DNS finds an address for the host

    The host name may be answered from a cache or resolved to one or more IP addresses.

    shop.example → A and/or AAAA records

  3. 03

    The browser establishes a secure connection

    The browser connects to an address and negotiates a transport and security for HTTPS.

    shop.example:8443

  4. 04

    The browser sends an HTTP request

    The request carries a method, target, and fields; the fragment is left out.

    GET /products?color=green HTTP/1.1 Host: shop.example:8443 Accept: text/html

  5. 05

    A server or intermediary handles the request

    A cache, CDN, proxy, or application may handle the request before a response is chosen.

    Browser → CDN cache → application → data store

  6. 06

    The response comes back

    A status code, response fields, and possibly content tell the browser what happened.

    200 OK Content-Type: text/html <!doctype html>…

  7. 07

    The browser builds and renders the page

    As HTML arrives, the browser parses it and coordinates style, layout, paint, and more requests.

    HTML → document tree + CSS + scripts + images